Lack of correlation between immunohistochemical expression of E2F-1, thymidylate synthase expression and clinical response to 5-fluorouracil in advanced colorectal cancer

Belvedere, O.
Puglisi, F.
Di Loreto, C.
Cataldi, P.
Guglielmi, A.
Aschele, C.
Sobrero, A.

¹Division of Medical Oncology, University of Udine, P. le S. M. Misericordia, Udine, Italy
²Department of Pathology, University of Udine, P. le S. M. Misericordia, Udine, Italy
³Division of Medical Oncology, S. Martino General Hospital, L. go R. Benzi 10, Genoa, Italy
⁴Division of Medical Oncology, Galliera General Hospital, Mura delle Cappuccine 14, Genoa, Italy

Received 25 June 2003; accepted 3 September 2003

^*Correspondence to: Prof. A. Sobrero, Medical Oncology, S. Martino Hospital, Largo R. Benzi, 10, 16132 Genoa, Italy. Tel: +39-010-555-3977; Fax: +39-010-555-5141; E-mail: [email protected]

Annals of Oncology 15(1):p 55-58, January 2004.

Background

The level of the enzyme thymidylate synthase (TS) is known to inversely correlate with the clinical activity of 5-fluorouracil (FU) in advanced colorectal cancer patients. Since the correlation is not very strong, we have retrospectively analyzed the expression of E2F-1 in tumor samples or metastases from 25 patients with advanced colorectal cancer, homogeneously treated with an FU-based regimen. E2F-1 is a transcription factor regulating the expression of TS along with other crucial DNA synthesis related enzymes.

Materials and methods

E2F-1 expression was analyzed by immunohistochemistry using the anti-E2F-1 monoclonal antibody KH95, scoring 2000 cells/case. Expression of TS was evaluated by immunohistochemistry using a rabbit anti-human polyclonal antibody.

Results

The level of E2F-1 expression did not correlate with TS expression, although a trend for correlation between E2F-1 level and maximal tumor shrinkage was observed (r=0.42; P=0.054).

Conclusions

In spite of previous reports demonstrating that E2F-1 quantified by rt-PCR and western blot correlates with TS and could be used as a predictor to select colorectal cancer patients more likely to respond to FU treatment, our data suggest that, under these experimental conditions, immunohistochemistry cannot be used for such selection.